Impact of the 2025 California Wildfires on Mountain Lion ( <i>Puma concolor</i> ) Habitat: A Remote Sensing and Species Distribution Modeling Framework

Abstract. Increasing wildfire frequency and intensity in southern California profoundly impacts ecosystems and keystone species like the mountain lion (Puma concolor), which relies on extensive and connected habitats. Quantifying the immediate ecological impact of the January 2025 Los Angeles County wildfires on Puma concolor habitat is a critical conservation concern. This study presents a novel, rapid-assessment framework for post-fire habitat evaluation, integrating Sentinel-2 burn metrics and VIIRS anthropogenic proxies via Google Earth Engine (GEE) with MaxEnt species distribution modeling (SDM). Comparative pre- and post-fire habitat suitability models were developed by integrating static environmental variables (topography, bioclimatic) with dynamic indices (Sentinel-2 NDVI, NBR quantifying vegetation response/burn severity; VIIRS nighttime lights quantifying anthropogenic presence) processed in GEE. To isolate the impact of fire-induced environmental changes on habitat suitability, the same set of pre-fire Puma concolor occurrence data from GBIF was used for both models. The results reveal a substantial habitat suitability loss of 104.12 km² (54.65%) within fire perimeters, with major losses concentrated in the Palisades, Eaton, and Hughes incidents. Model analysis confirms precipitation patterns (BIO16, BIO18) and temperature seasonality (BIO4) are major habitat drivers, but dynamic factors also played a key role, with post-fire vegetation condition emerging as a critically important factor influencing suitability, highlighting fire's impact on landscape resources. This underscores the necessity of incorporating dynamic, disturbance-specific RS data into SDMs for robust ecological impact assessment and highlights urgent conservation needs for Puma concolor in increasingly fire-prone urban-wildland interfaces.